Keywords - Function groups - @ A B C D E F G H I J K L M N O P Q R S T U V W X Y Z
Library: VaR
See also: VaRcdfDG cdf2quant
Quantlet: VaRqDG
Description: computes the a-quantile for the class of quadratic forms of Gaussian vectors; uses Fourier inversion to approximate the cumulative distribution function (CDF).
Usage: q = VaRqDG(a,par,N,K,dt)
Input:
a scalar, representing the probability level
par a list defining the distribution; contains at least the components: theta - the constant term delta - the linear term lambda - the diagonal of the quadratic term
N scalar, modulus of the FFT used in the inversion, should have a power of 2
K scalar, number of characteristic function evaluations to use, K <= N
dt scalar, grid-size in t to use for the approximation of the inversion integral
Output:
q scalar, the a-quantile
Example:
library("VaR")
theta = 0
delta = #(0)
lambda = #(1.4142)
par = list(theta,delta,lambda)
VaRqDG(0.95,par,512,512,0.1)
qfn(0.95)
Result:
Approximates a standard normal quantile and compares it
with the built-in function:

Contents of q
[1,]   2.7218

Contents of qfn
[1,]   1.6449
Author: S. Jaschke, 20011026 license MD*Tech
(C) MD*TECH Method and Data Technologies, 05.02.2006